An experiment on opinion formation: the role of objective information

How to cite

MORAES, FLAVIO; HEMSLEY, PEDRO; PAVÃO, LYNDA. 2022. "An experiment on opinion formation: the role of objective information." AEA RCT Registry. AEARCTR-0010131. https://doi.org/10.1257/rct.10131-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

FLAVIO MORAES* , PEDRO HEMSLEY , LYNDA PAVÃO

* Primary investigator

Registry ID
AEARCTR-0010131
DOI
https://doi.org/10.1257/rct.10131-1.0
First registered
Last updated
Status
In development
Trial dates
2022-09-27 – 2022-11-30
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/10131

Abstract

n this online experiment, we investigate the relationship between mental health and opinion formation - specifically, we evaluate opinions about Covid-19. We set up the experiment on Qualtrics and hire around 600 respondents on Amazon Mechanical Turk, restricting to the United States. In the first stage, we collect basic socio demographic information about the subjects. In the second stage, we run a questionnaire to classify subjects either as psychopaths or non-psychopaths, and then split the sample in these two groups (in fact, there is a third group of subjects that do not belong to these two first groups due to limitations in the classification system - this is not relevant to the questions we will pose). Each group then undergoes two randomizations.  In the first one, the treatment group receives objective information about the Covid-19 pandemic, while the control group receives no such information. In the second randomization, the treatment group answers a questionnaire about their feelings - the aim is to allow them to express emotions freely, in a process known as affect labeling. The control group does not answer this questionnaire. Finally, all subjects answer questions about the Covid-19 pandemic. We try to answer the following questions: 1- Psychopaths have different opinions about Covid-19?; 2- Psychopaths use objective information differently?; 3- Psychopaths respond differently to affect labeling?; 4- Affect labeling affects the impact of object information on opinion formation, maybe differently in each group (psychopaths or non-psychopaths)?

Experimental design

We set up the experiment on Qualtrics and hire around 600 respondents on Amazon Mechanical Turk, restricting to the United States.

Design details

Randomization method
Randomization done by on Amazon Mechanical Turk based on a form of socio demographic characteristics
Randomization unit
Individual
Clusters
1
Observations
600 individuals
Arms
300 individuals treatment, 300 individuals control
Primary outcomes
Opinions about Covid-19

Cited by

No citation data available for this registration yet.

Keywords